Proc. Natl. Acad. Sci. USA
Vol. 87, pp. 7713-7716, October 1990
Botany
Interplant communication: Airborne methyl jasmonate induces
synthesis of proteinase inhibitors in plant leaves
(jasmonic acid/pathogen/wound-inducible genes/localized/systemic defense responses)
EDWARD E. FARMER AND CLARENCE A. RYAN
Institute of Biological Chemistry, Washington State University, Pullman, WA 99164-6340
Contributed by Clarence A. Ryan, July 13, 1990
Inducible defensive responses in plants are
ABSTRACT
known to be activated locally and systemically by signaling
molecules that are produced at sites of pathogen or insect
attacks, but only one chemical signal, ethylene, is known to
travel through the atmosphere to activate plant defensive
genes. Methyl jasmonate, a common plant secondary compound, when applied to surfaces of tomato plants, induces the
synthesis of defensive proteinase inhibitor proteins in the
treated plants and in nearby plants as well. The presence of
methyl jasmonate in the atmosphere of chambers containing
plants from three species of two families, Solanaceae and
Fabaceae, results in the accumulation of proteinase inhibitors
in leaves of all three species. When sagebrush, Artemisia
tridentata, a plant shown to possess methyl jasmonate in leaf
surface structures, is incubated in chambers with tomato
plants, proteinase inhibitor accumulation is induced in the
tomato leaves, demonstrating that interplant communication
can occur from leaves of one species of plant to leaves of another
species to activate the expression of defensive genes.
FIG. 1. Structure of methyl jasmonate.
elicitors have been associated with membrane receptors (20)
and with changes in protein phosphorylation patterns of
membranes (21) and cellular proteins (22) during the induction process.
In this communication we report that a lipid-derived molecule, methyl jasmonate (Fig. 1), can act as a volatile signal
that induces the accumulation of proteinase inhibitor proteins
to even higher levels than can be induced by wounding. We
also report that the presence of methyl jasmonate from
sagebrush leaves can induce proteinase inhibitor I and II
accumulation in leaves of nearby tomato plants through the
atmosphere. Methyl jasmonate or its deesterified derivative,
jasmonic acid, is hypothesized to be a possible key component of intracellular signaling in response to wounding or
pathogenic attacks.
Activation of defensive genes in plants by pathogen and
herbivore attacks, or by other mechanical wounding, can
result from the action of a variety of signaling molecules that
are released in complex temporal patterns following the initial
invasion of the tissues (1-3). These signals are transported
locally by diffusion through intercellular and extracellular
fluids that permeate wound or infection sites or systemically
through the vascular system of the plants (4-7). Limited
indirect evidence has indicated that defense responses may
also be mediated by signals transported through the atmosphere (8-10). Whereas several chemicals, including ethylene, have been identified as candidate extracellular signaling
molecules for inducible defense genes, no direct biochemical
evidence has been presented that would implicate any volatile chemicals aside from ethylene as signals that can activate
plant defensive genes.
Among the defensive chemicals that are synthesized in
response to either herbivore or pathogen attacks are proteinase inhibitor proteins. Members of three families of woundinducible proteinase inhibitors, inhibitors I and II from Solanaceae and alfalfa trypsin inhibitor from Fabaceae, and
their cDNAs and genes, have been extensively characterized
(11-18). Proteinase inhibitors I and II are regulated at the
transcriptional level in response to wounding (19). We have
been investigating the signal transduction pathways that
regulate the synthesis of these inhibitors in response to
herbivore attacks and have found that the inhibitor genes are
induced by oligosaccharide fragments from plant (oligouronides) and pathogen (chitosan) cell walls (3). The mechanism by which oligosaccharide molecules activate the proteinase inhibitor genes is not known, but oligosaccharide
METHODS AND MATERIALS
Exposure of Plants to Methyl Jasmonate. Tomato (Lycopersicom esculentum cultivar Castlemart II) plants, 13-15
days after planting, were sprayed with solutions of 0.125%
(vol/vol) Triton X-100 with or without (±)-methyl jasmonate
(Bedoukian Research, Danbury, CT) or exposed to methyl
jasmonate vapor in 1250-ml air-tight glass chambers by
incubating plants together with cotton-tipped wooden dowels
to which had been applied 1 1.l of dilutions of (±)-methyl
jasmonate in ethanol or ethanol alone as a control. The cotton
tip was placed -4-6 cm from the plant leaves. The chambers
were incubated in constant light (300 microeinsteins-m-2.
sec-1) at 280C for 24 h. Cuttings, in 10 ml of water, from
2-month-old tobacco plants (cultivar Xanthi) and from 1month-old alfalfa plants (line RA3) were exposed to methyl
jasmonate vapor in air-tight glass chambers in a similar way.
Leaf juice was expressed from the leaves and assayed for
proteinase inhibitors I and II by radial immunodiffusion (23,
24).
Purification and Characterization of Methyl Jasmonate from
Artemisia tridentata Nutt. ssp. tridentata. The terminal 15 cm of
A. tridentata branches were collected from a natural population growing at Lyons Ferry, WA. Branches (1 kg) containing several hundred small leaves were agitated in ethanol
(1 liter) for 10 sec. The resultant mixture was concentrated to
a brown oil (2.4 ml) that was extracted into pentane (75 ml).
The pentane was removed by rotary evaporation and the
resultant yellow oil (1.2 ml) was fractionated on a silica gel
The publication costs of this article were defrayed in part by page charge
payment. This article must therefore be hereby marked ''advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
7713
7714
Botany: Farmer and Ryan
(Silicar, 60A; Mallinckrodt) column (1.2 x 20 cm). The
column was first washed in pentane. The ethyl acetate
content of the pentane was increased in 5% (vol/vol) steps to
10% (vol/vol) and then increased in 10% steps to 100% ethyl
acetate. A fraction containing volatile proteinase inhibitorinducing activity eluted in 30% (vol/vol) ethyl acetate. This
fraction was further resolved by preparative thin-layer chromatography on silica gel, developed in benzene/ethyl acetate, 10:1 (vol/vol). Compounds migrating with an Rf of 0.38
exhibited proteinase inhibitor-inducing activity when exposed to young tomato plants. This active material was
further chromatographed by analytical silica gel thin layer
(Kodak) chromatography. The thin layer sheet was presoaked in 5% (wt/vol) silver nitrate in 75% (vol/vol) methanol. Before chromatography the plate was allowed to dry for
20 h at 28TC. The plate was chromatographed in 30% (vol/vol)
ethyl acetate in hexane. Proteinase inhibitor-inducing activity was found associated with components that migrated at an
Rf of 0. 17. This material was further resolved isocratically by
reverse-phase HPLC on a Beckman Ultrasphere C18 ion pair
column (4.6 x 250 mm, 5 ,um) in 50% acetonitrile/0.1%
trifluoroacetic acid in water. A single fraction containing a
partially resolved double peak copurified with the proteinase
inhibitor-inducing activity. Gas chromatography/mass spectroscopy was carried out on a Hewlett- Packard 5985 GC/MS
system. The column was a Superox FA (30 m x 0.25 mm;
Alltech Associates) with He carrier gas at 1.4 kg'cm-2. The
column temperature was 450C for 5 min and then was
increased at 10'C/min to 220'C, which was held for 10 min.
RESULTS AND DISCUSSION
Methyl jasmonate is a naturally occurring compound that has
been identified in plants from at least nine families (25) and
has been utilized as a perfume fragrance for decades. This
compound has recently been the subject of considerable
interest because of its biological activities in plants in inducing senescence and in regulating vegetative storage proteins
in plant leaves (25). The proteinase inhibitor-inducing ability
of methyl jasmonate was first noted when, upon spraying the
compound on leaves of tomato plants, it powerfully induced
the synthesis and accumulation of proteinase inhibitor I
protein (Table 1). The chemical induced the accumulation of
inhibitor I to levels higher than could be induced by wounding. It was also noted during these experiments that control
plants that had not been sprayed with methyl jasmonate, but
incubated in the same chambers with the sprayed plants,
accumulated low levels of proteinase inhibitor I protein
(Table 1). Control plants incubated in separate chambers did
not accumulate inhibitor I at all. These reproducible results
Table 1. Accumulation of proteinase inhibitor I in tomato leaves
induced by methyl jasmonate or wounding
Proteinase inhibitor I,
n
Plants
tig/g of tissue
Chamber A*
210 ± 7
6
Methyl jasmonate-sprayed
28 ± 3
6
Control
Chamber Bt
86 ± 8
4
Wounded
0
6
Control
Proteinase inhibitor I levels were assayed by radial immunodiffusion (22, 23). Values are reported as mean ± SEM.
*Sprayed and control plants were allowed to dry in the open at room
temperature for 30 min and then placed in large sealed Plexiglas
boxes (capacity, 11.34 liters) and incubated under constant light
(300 microeinsteins m-2 sec-1) for 24 h.
tPlants were wounded once across their lower leaves with a hemostat
and assayed 24 h later.
Proc. Natl. Acad. Sci. USA 87 (1990)
suggested that volatile methyl jasmonate was inducing the
synthesis of proteinase inhibitors in the nearby untreated
control plants. To further test this possibility, tomato plants
were placed in air-tight chambers together with cotton-tipped
wooden dowels onto which various dilutions of methyl jasmonate in ethanol had been applied to the cotton. The dowels
were placed so that no interaction was possible between the
methyl jasmonate and the plants except through the atmosphere. After incubating the plants in light for 24 h following
the introduction of methyl jasmonate to the chambers, leaf
juice from the plants was assayed for inhibitor I and II protein
levels. The presence of increasing levels of methyl jasmonate
in the different chambers resulted in the synthesis and
accumulation of proteinase inhibitors I and II in a dosedependent manner (Fig, 2). Below about 10 nl of methyl
jasmonate per chamber the response was dose dependent.
The results in Fig. 2 suggested that at low concentrations of
methyl jasmonate on the cotton wicks the volatile levels were
limiting and that leaves were not able to assimilate enough of
the molecules to maximize the signaling response. Above 10
nl per chamber, volatile methyl jasmonate appeared to be at
high enough concentrations in the chambers to be assimilated
at levels that could maximally induce the inhibitors.
When enough methyl jasmonate to cause near-maximal
induction of proteinase inhibitors was present in the chambers (100 nl per chamber), tomato plants began to accumulate
proteinase inhibitors I and II at about 5 h after the initial
exposure to the volatile compound and continued to accumulate the proteins at linear rates for nearly 20 h (Fig. 3).
After about 20 h, however, the rates of accumulation of the
two inhibitors declined. The reasons for this decline are
unclear. In wounded plants the rates of accumulation of the
proteinase inhibitors declined about 10 h after an initial
wound, but the wound induction could be reinforced by a
wound administered about 12 h later (19). The kinetics of
uptake of methyl jasmonate and its fate in the leaves await
further investigation.
Exposure of tomato plants for only 30 min to gaseous
methyl jasmonate originating from 100 nl of the chemical per
chamber was sufficient to induce a moderate level of accumulation of proteinase inhibitor protein (Fig. 4). With increasing times of exposure to the volatile methyl jasmonate
the accumulation of inhibitor protein increased and approached a maximum at about 8 h.
To determine if methyljasmonate could activate proteinase
inhibitor synthesis in other plant genera and another plant
0)
40
200-
200-
o
4-10
0
0..
0
0.1
1
10
100
1o00
Methyl Jasmonate (nL/assay)
FIG. 2. Induction of proteinase inhibitor I and II proteins in the
leaves of intact tomato plants after exposure to airborne methyl
jasmonate for 24 h. Two-week-old tomato plants were placed in
1250-ml air-tight glass chambers together with cotton-tipped wooden
dowels to which had been applied 1 ,ul of dilutions of (±)-methyl
jasmonate in ethanol. The cotton tip was placed -4-6 cm from the
plant leaves. The jars were incubated in constant light at 28°C for 24
h. Leaf juice was analyzed for proteinase inhibitors I and II. B3ars
indicate the SEM (n = 4). o, Inhibitor I; e, inhibitor II.
Botany: Farmer and Ryan
Proc. Natl. Acad. Sci. USA 87 (1990)
0300-
2-
0
c1
100
000
10
20
Time
40
(h)
FIG. 3. Time course of induction of proteinase inhibitor I and II
proteins in the leaves of tomato plants in response to airborne methyl
jasmonate. Tomato plants were incubated in air-tight glass jars
together with a cotton tipped dowel on which 1 Al of a solution of 100
nl of methyl jasmonate in 1 ,ul ethanol had been applied. At various
times after the initial application of methyl jasmonate, levels of
proteinase inhibitors I and II were determined. Bars show the SEM
(n = 4). o, Inhibitor I; e, inhibitor II.
family, small tobacco plants (Xanthi) and small alfalfa plants
(line RA3) were exposed to volatile methyl jasmonate (100 nl
of methyl jasmonate per 1250-ml chamber) and leafjuice was
assayed for the presence of tobacco trypsin inhibitor and
alfalfa trypsin inhibitor. In the presence of methyl jasmonate,
tobacco trypsin inhibitor levels were elevated from 7 + 3
,ug/g of tissue (n = 4) in leaves of control plants to 116 37
,g/g of tissue (n = 4) in leaves of plants exposed to methyl
jasmonate. Alfalfa trypsin inhibitor levels were elevated from
33 7.3 ,g/g of tissue (n = 4) in control plants to 385 26.5
,ug/g of tissue (n = 4) in exposed plants. Thus airborne methyl
jasmonate induced the expression of several proteinase inhibitor genes representing three inhibitor families in leaves of
plants from the Solanaceae and Fabaceae families.
Because of the widespread occurrence of methyl jasmonate in the plant kingdom, we investigated the possibility
that a species of plant that contained methyl jasmonate in its
leaves could induce expression of proteinase inhibitor genes
in nearby tomato plants. If so, it would establish a biochemical basis for a previously unrecognized form of defense gene
±
±
7715
regulation involving interplant communication through the
atmosphere. At least one member of the genus Artemisia,
Artemisia absinthium, is known to contain methyl jasmonate
in its leaves (26). We therefore chose a related species, the
sagebrush A. tridentata Nutt. ssp. tridentata, an ecologically
dominant species in the western United States, as a possible
donor of volatile methyljasmonate. Small tomato plants were
incubated in air-tight chambers in the absence or presence of
5 g of leafy branches of A. tridentata, with no direct physical
contact at any time between the plants. Within 2 days the
leaves of the tomato plants exhibited elevated levels of
proteinase inhibitors I and II (Table 2). These experiments
(assaying the induction of inhibitor I) have been repeated
numerous times. Although individual experiments vary
somewhat, the results-i.e., the induction of inhibitor I-was
always noted.
To identify the volatile signaling molecule, cotton swabs
were brushed across the leaf surfaces of A. tridentata leaves
and placed in glass chambers with tomato plants, similar to
the experiments described previously with methyl jasmonate. A volatile component was present that induced the
accumulation of proteinase inhibitors in the leaves (data not
shown). To determine if methyl jasmonate was actually
present in the leaf surface trichomes of A. tridentata, an
ethanolic fraction from the leaf surface was obtained and
fractionated into its components by silica gel column chromatography, thin-layer chromatography, and reverse-phase
HPLC. At each step of purification, components were assayed for volatile proteinase inhibitor-inducing activity. Active fractions were further purified. Reverse-phase HPLC of
a biologically active fraction from silver nitrate thin-layer
chromatography produced two overlapping peaks with retention times similar to two methyl jasmonate isomers. The
mass spectrum of one of the isomers isolated from A.
tridentata is shown in Fig. 5 to be virtually identical to a
sample of chemically synthesized methyl jasmonate isomer.
These results strongly suggest that the airborne chemical
signal from A. tridentata leaves that induces the expression
of the proteinase inhibitor genes is methyl jasmonate.
The chemical structure of jasmonic acid is similar to the
prostaglandins, important signaling molecules in animals
(25). Jasmonic acid is apparently synthesized from linolenic
acid, a fatty acid ubiquitous in plants (27). The release of
linolenic acid, triggered by the activation of specific lipases
in response to pest or pathogen attacks, could lead rapidly to
300
*D
CD
2
100
0
00
2
4
6
Time of Exposure
8
10
(h)
FIG. 4. Synthesis of proteinase inhibitor I protein in response to
increasing times of exposure of tomato plants to airborne methyl
jasmonate. Tomato plants were incubated in air-tight 1250-ml glass
chambers in the presence of volatile methyl jasmonate (100 nl
dissolved in 1 ,ul of ethanol) pipetted onto cotton swabs. After
exposure to gaseous methyl jasmonate, the plants were transferred
to chambers free of methyl jasmonate and further incubated for a
total of 24 h after the initial exposure, when leaf juice was assayed
for proteinase inhibitor I content. Bars show the SEM (n = 4).
Table 2. Induction of proteinase inhibitors I and II in leaves of
tomato plants incubated in the presence of leafy branches of the
sagebrush A. tridentata ssp. tridentata
Proteinase inhibitor
I
II
Plants
/Lg/g of tissue
n
tkg/g of tissue
n
Experiment 1
Incubated with
A. tridentata
54.5 + 8.7
4
ND
Control
0
4
ND
Experiment 2
Incubated with
A. tridentata
71.0 + 16.1
10
50.6 + 13.7
12
Control
0
10
0
12
The plants were placed so that no physical contact was possible
between the sagebrush leaves and tomato leaves. The chambers
(volume, 1250 ml) were sealed, incubated for 2 days under the
conditions described in Table 1, and assayed. Control tomato plants
were placed in separate chambers in the absence of sagebrush and
incubated under identical conditions. Plants for experiment 1 were
collected from 1 mile (=1.6 km) east of Desert Aire, WA (location I).
Plants for experiment 2 were collected from Lyons Ferry, WA
(location 2). ND, not determined.
7716
Botany: Farmer and Ryan
Proc. Natl. Acad. Sci. USA 87 (1990)
100-3
significance that has been seldom considered in evaluating
interactions within and between plant communities.
Methyl Jasmonate
80-
95
20-
i20-
141
87
L316
121
151
167 17 193 20 24
l22Ll
1'7
19'
Li2i0r8
ir, 177
i 14o4
60
40
80,100
120
140
160
180
200
220
A.ijdentnata. extract
100-
We thank V. Franceschi for a gift of methyl jasmonate, A. Koepp
and R. Croteau for assistance with GC/MS analyses and Greg
Wicheins for growing plants. Correct identification of Artemisia was
confirmed by J. Mastrogiuseppe, Owenby Herbarium (Washington
State University). This work was supported in part by Washington
State University College of Agriculture and Home Economics Project 1791, National Science Foundation Grant DCB-8702538, the
McKnight Foundation, and EniMont Americas, Inc.
80-
60-
~~~~151
96
20-
103
400
s6800
121 133
'10
1617193
160
180
200
224
220
FIG. 5. Mass spectrum of authentic methyl jasmonate compared
to methyl jasmonate isolated from the sagebrush A. tridentata ssp.
tridentata.
the production of jasmonic acid through the action of cyclooxygenases. The possible role of jasmonic acid in signal
transduction pathways leading to localized and systemic
defensive gene expression needs to be investigated.
Airborne methyl jasmonate molecules may enter the vascular system by way of stomates and activate the proteinase
inhibitor genes through a receptor-mediated signal transduction pathway. Alternatively, methyl jasmonate may diffuse
into the leaf cell cytoplasm where it would be hydrolyzed to
jasmonic acid by intracellular esterases. The free acid may,
in turn, be an integral part of a general signal transduction
system that regulates inducible defensive genes in plants.
Jasmonic acid has a much lower vapor pressure than methyl
jasmonate, but it does induce proteinase inhibitors in tomato
leaves, albeit more weakly, when present in closed chambers
on cotton wicks (data not shown). We have supplied solutions of jasmonic acid to young tomato plants through their
cut petioles and this compound induces proteinase inhibitor
synthesis, but it must be supplied at micromolar levels. It
may be that the free jasmonic acid does not easily penetrate
cells. This aspect of the signaling phenomenon needs further
study.
Previous studies have shown that methyl jasmonate, or
jasmonic acid, when applied directly to plants can produce
various responses, including growth inhibition (28, 29), promotion of senescence and/or abscission (26, 30), and induction of specific leaf proteins in monocots and dicots (31-33).
The data herein demonstrate that a highly sensitive mechanism is present in Solanaceae and Fabaceae families that can
activate proteinase inhibitor genes in response to volatile
methyl jasmonate. The mechanism may be broadly present in
nature. Whether volatile methyl jasmonate can activate the
other responses mentioned above is not known and should be
addressed. It is possible that methyl jasmonate, or other
volatile signals, released by such plants as A. tridentata, have
multiple effects on nearby plants, either by inducing the
expression of defensive genes, or genes involved in other
responses such as senescence, in distal tissues of the same
plants or in neighboring plant species. If such signaling is
widespread in nature it could have profound ecological
1. Darvill, A. G. & Albersheim, P. (1984) Annu. Rev. Plant
Physiol. 35, 243-275.
2. Lawton, M. A. & Lamb, C. J. (1987) Mol. Cell. Biol. 7,
335-341.
3. Ryan, C. A. (1987) Annu. Rev. Cell Biol. 3, 295-317.
4. Green, T. R. & Ryan, C. A. (1972) Science 175, 776-777.
5. Kuc, J. & Preisig, C. (1984) Mycologia 76, 767-784.
6. Kopp, M., Rouster, J., Fritig, B., Darvill, A. & Albersheim, P.
(1990) Plant Physiol. 90, 208-216.
7. Hammond-Kosack, K. E., Atkinson, H. J. & Bowles, D. J.
(1989) Physiol. Mol. Plant Pathol. 35, 495-506.
8. Baldwin, I. T. & Schultz, J. C. (1983) Science 221, 277-279.
9. Rhoades, D. F. (1983) in Plant Resistance to Insects, ed.
Hedin, P. (Am. Chem. Soc., Washington, DC), pp. 55-68.
10. Zeringue, H. J. (1987) Phytochemistry 26, 1357-1360.
11. Melville, J. C. & Ryan, C. A. (1972) J. Biol. Chem. 247,
3445-3453.
12. Bryant, J., Green, T., Gurusaddaiah, T. & Ryan, C. A. (1976)
Biochemistry 15, 3418-3423.
13. Brown, W. E. & Ryan, C. A. (1984) Biochemistry 23, 34183422.
14. Graham, J. S., Pearce, G., Merryweather, J., Titani, K., Ericsson, L. & Ryan, C. A. (1985) J. Biol. Chem. 260, 6555-6560.
15. Graham, J. S., Pearce, G., Merryweather, J., Titani, K., Ericsson, L. & Ryan, C. A. (1985) J. Biol. Chem. 260, 6561-6564.
16. Lee, J. S., Brown, W. E., Graham, J. S., Pearce, G., Fox,
E. A., Dreher, T. W., Ahern, K. G., Pearson, G. D. & Ryan,
C. A. (1986) Proc. Natl. Acad. Sci. USA 83, 7277-7281.
17. Thornburg, R. W., An, G., Cleveland, T. E., Johnson, R. &
Ryan, C. A. (1987) Proc. Natl. Acad. Sci. USA 84, 744-748.
18. Sanchez-Serrano, J. J., Keil, M., O'Connor, A., Schell, J. &
Wilmitzer, L. (1987) EMBO J. 6, 303-306.
19. Graham, J. S., Hall, G., Pearce, G. & Ryan, C. A. (1986)
Planta 169, 399-405.
20. Schmidt, W. E. & Ebel, J. (1987) Proc. Natl. Acad. Sci. USA
84, 4117-4121.
21. Farmer, E. E., Pearce, G. & Ryan, C. A. (1989) Proc. Natl.
Acad. Sci. USA 86, 1539-1540.
22. Dietrich, A., Mayers, J. E. & Hahlbrock, K. (1990) J. Biol.
Chem. 256, 6360-6368.
23. Ryan, C. A. (1967) Anal. Biochem. 19, 424-440.
24. Trautman, R., Cowan, K. M. & Wagner, G. G. (1971) Immunochemistry 8, 901-916.
25. Anderson, J. M. (1989) in Second Messengers in Plant Growth
and Development, ed. Boss, W. F. & Morre, D. J. (Liss, New
York), pp. 181-212.
26. Ueda, J. & Kato, J. (1980) Plant Physiol. 66, 246-249.
27. Vick, B. A. & Zimmerman, D. C. (1984) Plant Physiol. 75,
458-461.
28. Dalthe, W., Roensch, H., Preiss, A., Schade, W., Sembdner,
G. & Schreiber, K. (1981) Planta 153, 530-535.
29. Yamane, H., Takagi, H., Abe, H., Yokota, T. & Takahashi, N.
(1981) Plant Cell Physiol. 22, 689-697.
30. Curtis, R. (1984) Plant Growth Regul. 3, 157-168.
31. Mueller-Uri, J., Parthier, B. & Nover, L. (1988) Planta 176,
241-247.
32. Anderson, J. M., Spilatro, S. R., Klauer, S. F. & Franceschi,
V. R. (1990) Plant Sci. 62, 45-52.
33. Staswick, P. E. (1989) Plant Cell 2, 1-6.